I still see that, wow, good morning from sunny Victoria, Australia. So someone from Australia is joining, Belinda. So we're going to have a quite an international group here. Okay, I'm going to wait for 30 more seconds because I can see that still the participant audience is coming in. So we'll, UK, there we go. All right, Dr. Polsky and Dr. Willen, be ready for those international questions then. I see Scotland, wonderful morning to you guys. Okay, it's 3.31 on my screen, that's in mountain time, and I'm going to now correct my introduction here and say good morning to folks from UK, Scotland, Australia, other part of the world. Good evening for the East Coast people and maybe late afternoon for the West Coast. We appreciate your time. I understand this is a very interesting topic, the topic that I think many times we struggle about what to do during pregnancy, complicated by diabetes, do we need to use this closed loop system? Are they really made for pregnancy? So we have two wonderful speakers to really answer that question. And probably it will give you some perspective about if you want to use the hybrid closed loop system during pregnancy in people with diabetes. This webinar is made possible with a joint effort of Diabetes Technology Interest Group and then ADA's Pregnancy and Diabetes Interest Group. And those who are wondering about what is this interest group, then just Google ADA Interest Group. It will take you to the professional.diabetes.org website where you can be a part of the interest group and learn from each other. So about our speakers, Dr. Sarit Polsky, I have worked with her for a long period of time here at the Barbara Davis Center for Diabetes in Aurora, Colorado. She's an associate professor, adult endocrinologist, and Dr. Amy Valant from OHSU in Portland, Oregon, and she's an associate professor of OB-NGY and maternal fetal medicine there. So you will get a perspective from an endocrinologist and an OB-GY MFM. So it will be a very interesting webinar that I'm looking forward to. So there are a couple of announcements before we'll start. For folks, we know that there are plenty of different kind of CGMs and pump, and it's sometimes overwhelming for all of us how to make those technology work for our clinic. There is a really great, interesting on-demand webinar available, and so there is a registration link on the bottom of this slide. Please register for that. I'm sure it will be a lot of learning in this webinar. ADA is also—oops, I am forwarding it faster—ADA has also pretty interesting modules. It's in development right now, but it's coming soon. It will be open up by March. It's called Making Diabetes Technology Work. It's a different 14 topics from understanding of these technologies to implementing them into your clinical practice to making the best of the technologies to improve outcome. And I see that Jennifer and Caitlin from ADA are putting those registration links on the chat field, so you can click those links and learn more about that. You can do a preregistration right now. It's coming up. It's a pretty exciting program from the ADA. There are a lot of other webinars by different ADA's interest groups. That includes the epidemiology, like, for example, diabetes risk score. And tomorrow, we have other seminars, for example—webinars, for example, standard of care, using the AID system in kids, and so on. So again, go to this website, professional.diabetes.org forward slash webinars, and you will learn a lot about different kind of opportunities to attend those webinars and learn from each other. With that, I am going to hand over this program to Dr. Sarit Polsky and Dr. Amy Vallant. Now the stage is yours, Dr. Polsky. Great. Thank you. Dr. Vallant and I would like to thank Dr. Shah and the organizing committee for inviting us to speak on this really exciting topic of hybrid closed-loop therapy and pregnancies associated with diabetes. Here are our dualities of interest, and we both have received research funding from different industry partners to study advanced diabetes technologies in pregnancies managed in women who have diabetes. So first, I'm going to give you a very brief background about hybrid closed-loop therapy and pregnancy. So all the next few slides that I'm going to discuss were about an investigational device, so they were off-label use of a product. I'm trying to advance, there we go. While the nomenclature for automated insulin delivery systems has many different names, all the studies in pregnancy are hybrid closed-loop, meaning there has to be a meal announcement of carbohydrate intake for these automated insulin delivery systems. And of course, we all want to speak the same language in terms of what our goals are for glycemic levels. And so in the 2019 international consensus statement, of course, the recommended range outside of pregnancy for glucose control was 70 to 180 more than 70% of the time. In pregnancies associated with diabetes, we changed that range so that it's both lower and narrower to 63 to 140 milligrams per deciliter, or 3.5 to 7.8 millimoles per liter. For women with type 1 diabetes, there is a specific recommendation that we stay within that range more than 70% of the time. For gestational diabetes and type 2, it should be well above 70, but we don't know that exact amount. And then we're hoping for less than 4% of values under 63 and less than 25% over 140. So the initial work on this was done by Helen Murphy and Roman Havorka over in the United Kingdom, and it's work that really represents a decade of research. And what I'm showing you here is a very brief description of the closed-loop in pregnancy or CLIP series of studies. The first two, CLIP 1 and CLIP 2, were 24 hours of wear, supervised the entire time in a clinical research facility, and sort of proof of concept studies. CLIP 3 was an open-label, randomized crossover design. So women either went to sensor-augmented pump therapy or hybrid closed-loop therapy overnight and sensor-augmented pump therapy during the day for the same devices for four weeks of wear, had a two-week washout, and then four weeks of wear for the alternate randomization assignment. In CLIP 4, same study design, but in that case, they do 24 hours of hybrid closed-loop therapy or 24 hours of SAPT for four weeks, two weeks, and then the other assignment. So in CLIPS 1 and 2, what they saw initially is that it looked quite promising. And in fact, that time and range was really high, about 80 to 100%. Time below range was a bit higher in CLIP 2 than in CLIP 1, but these showed initially that it seemed to be safe and effective, and we could move on to some longer-term use and at-home use of these products. And they were also refined in between each study, so the investigational algorithm was adjusted after each study. Now for CLIPS 3 and 4, the algorithm was actually housed on a tablet computer and not on an insulin pump, just to show you kind of the iterations of how we move through research and then get to the point where it's then eventually housed within the pump. Here I'm showing you from CLIP 3, the 24 hours of data using closed-loop overnight and SAPT during the day compared to sensor-augmented pump therapy during the day. And what you can see in CLIP 3 is that the mean sensor glucose in the closed-loop arm was lower, 128 versus 137 in sensor-augmented pump therapy. The time spent in the optimal time and range was increased by about 10% in the closed-loop group, and the time spent in hyperglycemia was reduced, with no effect on hypoglycemia. Alternatively, another way to look at this is with the ambulatory glucose profile. Some picture is worth a thousand words. So in here you can see across the time period that orange, which is the closed-loop arm, has a lower sensor glucose value overnight, which is when hybrid closed-loop was engaged, compared to sensor-augmented pump therapy in gray, which has higher values overnight. In CLIP 4, this was a three-site study. They enrolled 16 women, and what they did is they had a two- to four-week run-in period, four weeks of the first randomization assignment, two-week washout, four weeks of the second, and then all of the women had the option to continue closed-loop for the rest of the study. And in fact, in CLIP 3, they also had the option to continue. And what we see is that, unlike CLIP 3, where we saw improvements in the mean sensor glucose and the time within range and then the time above range, there were no differences in CLIP 4 among those metrics. However, there were differences in terms of hypoglycemia, such that the closed-loop arm experienced less hypoglycemia and less risk for hypoglycemia, as depicted by the low blood glucose index, again, showing promise for hybrid closed-loop therapy in pregnancy. Additionally, there are some products that don't have a pregnancy-specific algorithm, as was used for those CLIP studies, but rather, these case series and case reports are showing about use of either commercially available products that don't have a pregnancy-specific algorithm, such as the Medtronic 670G, or DIY, do-it-yourself systems. And so there are these reports emerging that are showing, in general, that it seems to be working well for some women. So now what we're going to have is a little bit of a debate, and I'm on the side of potential advantages of closed-loop in pregnancies associated with diabetes. So one of the reasons why I think it can be advantageous to consider this therapy is because of this number, 45%. This is the number of unplanned pregnancies in the United States, hasn't changed that much over the last few years. As such, there are women using closed-loop therapies right now that are not approved for pregnancy, and they become pregnant while they're using them, and they therefore sometimes choose not to stop that therapy when they become pregnant. So it is something that we do have to deal with in our everyday practice on a regular basis. And I just want to take the opportunity to remind us all that if you care for a woman with a chronic disease, whether it be diabetes, thyroid disease, adrenal, whatever it is, that we should be discussing reproductive health issues multiple times over the reproductive health span and encouraging women to come in for preconception counseling. Now, of course, one of the things that is so important in pregnancy is managing the changes in the insulin requirements over the course of gestation. So in this graph, what you're seeing on the y-axis is qualitatively levels of insulin, normal, two times normal, and three times normal, and what you're seeing on the x-axis is where we are in gestation from conception to the 40-week estimated due date to postpartum. And initially, there's a drop in the insulin requirement because of insulin sensitivity in the first trimester. That drop may be 20% or 30% compared to pre-pregnancy doses, and then the insulin resistance of pregnancy ensues due to maternal and placental hormones that causes that insulin requirement to go up 200% or 300%, and then immediately after baby and placenta are delivered, those insulin requirements plummet, and they may be even less than that insulin-sensitive stage of the first trimester, maybe 40% less compared to pre-pregnancy doses. Women who have gestational diabetes don't have the endogenous insulin secretory capacity to overcome the hyperglycemia of insulin resistance in pregnancy. So I bring this to you to show you it's really difficult to keep up with these changes in the insulin requirements and to ask, can hybrid closed-loop therapy help in such a situation? So it's difficult to talk about closed-loop without talking about CGM, which is one of the most important aspects of a closed-loop system. So the landmark study here was the concept study. It was performed in 31 hospitals in the UK, Canada, and one U.S. site, and basically what they looked at was self-monitoring blood glucose compared to CGM. What they showed is that there was a difference and an improvement in the A1c from baseline to both 24 weeks and to 34 weeks for both groups. The CGM group and the self-monitoring blood glucose group. What they were hoping to achieve was a difference of 0.5% in the final A1c at 34 weeks. And in fact, they did not achieve that. In fact, the difference between the A1cs was 0.18% between groups. Despite the fact that in this relatively well-controlled population, the A1c values were pretty close together, there was a significant difference in the time spent in maternal glucose control in such that the time in range was about 7% higher in the CGM users, and the time in the hyperglycemic range was about 5% lower in the CGM users. So the question becomes, does this translate to any improvements in maternal or neonatal health? Well, not for maternal health, but significant improvements for neonatal health. So improvements in median birth centile, decreased risk for large for gestational age babies, NICU admissions over 24 hours, neonatal hyperglycemia requiring IV dextrose, and it reduced the infant length of stay by one day. So what do we see? In this first study of constant CGM use, the maternal glucose improvements derived from CGM use, meaning improvement in time and range and reduction in postprandial hyperglycemia led to improvements in neonatal health, but only when the CGM was used most of the time. So it had to be used about 70% of the time or more. Of interest as well is to show you that there were unfortunately still severe hypoglycemic events and diabetic ketoacidosis in both groups, CGM and SMBG. So the question becomes now, can hybrid closed-loop immunotherapy, can the algorithm that could stop insulin delivery in an impending low or increase insulin delivery for an impending severe high, improve some of these severe outcomes like hypoglycemia and DKA in pregnancy? That's that next layer of research. So I want to bring home the point that it's really difficult to manage the glucose levels throughout gestation. And unfortunately, we don't really get to where we want to get to for most women. So that concept study was combined in their data with Christensen et al.'s data from a Swedish study of two tertiary centers with observational data of CGM use using the Libre Index Com G4 systems. What they did is they took data from these 386 women. They examined weekly CGM metrics, that is, if there was data in four or more days within a given week, they looked at the CGM metrics for that particular week of gestation. And then they additionally plotted, you know, what is the risk for a large for gestational age infant based on the CGM metrics. Now, this first figure is showing you the mean glucose level from 0 to 10 to 20, 30 and 40 weeks of gestation. And then you can see on the y-axis in millimoles per liter, and I put an orange there, milligrams per liter, what that mean glucose translates to. Look at this pattern. In the beginning of pregnancy, women are starting off with a higher mean glucose than what we would recommend. There is a significant drop till about 10 weeks, then there's sort of a plateau and then a further improvement around 32 weeks or later. Now in the time and range, we want 70% time and range. We only get there in the final weeks of the pregnancy. In time above range, we only get there in the final weeks of the pregnancy. And that's where that green line is showing you, I added that to these figures, what our target is. And then hypoglycemia all over the board and not at goal at any time point. Just to bring home the point, with standard of care, all these women were using a CGM with or without a pump. We still didn't get to where we needed to be most of the time. So can closed loop therapy help to get us to where we need to be earlier and keep us there? There was a lot of individual variation in the CLIP4 study. What I'm showing you here is in the red dots, closed loop, green dots, sensor augmented pump therapy. And then the blue dot is that continuation phase where all 16 women opted to continue on closed loop. Now in participant number one, she started off with a high time and range. The time and range is on the Y axis. It's around 75%. Her green dot, the medium of those green dots, dips a little bit when she goes into sensor augmented pump therapy, and then it goes up again when she continues. So even in a woman who starts off with very high time and range, she does better with closed loop. In participant number two, she had a significant drop in her sensor augmented pump therapy data compared to when she was first randomized to closed loop. And then when she goes back on closed loop, her numbers increase again. So she did much better with closed loop. In participant number 10, she was first randomized to sensor augmented pump therapy. Her time and range is around, you know, 40%. It goes up when she first starts closed loop, and it continues to go up for the rest of gestation. There were no episodes of severe hypoglycemia or DKA during this trial. Again, small numbers, 16 women, but still showing promise. Now one of the interesting things they did for CLIP4 was they stratified women based on their booking or enrollment A1c. So they looked at women who came into the study with an A1c of 7.5% or less, and they compared glucose metrics to women who started off with an A1c over 7.5%. Now look at what happens here. The women who started off with a lower A1c to begin with, they do pretty well when they're in sensor augmented pump therapy and when they're in closed loop. Their closed loop time and range is a little bit higher, 72% versus 69%. The women who start off with a higher A1c don't really have an improvement compared to the other randomization assignment. When we look at time spent below range, no significant changes, and time spent in mean glucose, much better in the women who start off with a lower A1c to begin with. Now here's the interesting part of this slide. Well, there's lots of interesting parts, but here's one interesting part I want to show you. Look at the 36-week time point, final end point. Women who started off already with a time and range of around 70% at the beginning, still improved by about seven and a half, 8% in their time and range with closed loop therapy. Women who started off with a higher A1C to begin with and had a 57% time and range, improved more like 12% with their time and range. So this kind of suggests it can be helpful for all groups, but there is gonna be a difference if you start off with better control to begin with. I was fortunate enough to publish the first case series of the Medtronic 670G system with a colleague of mine, Dr. Ochturk. These were cases of three cases of women who used the Medtronic 670G in auto mode for some or all of their pregnancy. Now, you may ask why in the world would we even consider doing off-label use of this product in pregnancy, which has a target of 120 and we want fasting glucose levels under 95 for pregnancy. And the reason is if we look at clip three, clip four and the concept study, the mean glucose is already above 120 most of the time. And so my hypothesis was that for certain individuals, maybe it's still better even if we only get to a mean glucose of 120. This is a very busy slide. I'm gonna focus your attention where you'd like it to go. This is the case one. She started off already with an A1C in the range that we would prefer early pregnancy. And she was actually using the Medtronic 670G in manual mode at the beginning of pregnancy with very high levels of hypoglycemia. You can see 12 to 13% below 70. Around halfway through, we said, you know what? You're having a bunch of lows. Those lows are being over-corrected and leaning to rebound hyperglycemia. Let's see if we can put you in automation and smooth things out and get rid of all these lows. So in fact, that's what we did. Her mean sensor glucose initially went up with that insulin resistance of pregnancy and then it improves. And by the end, you can see that her mean glucose is around 125 to 114, similar to what it was at the beginning of pregnancy, but with a third to a fourth amount of hypoglycemia. The way that she was able to achieve this though, to have a mean glucose under that 120 target of the algorithm is by putting in fake carbohydrates, by assisting the system. And she also happened to do quite well during labor and delivery with auto mode. Now case two had much more difficulty controlling her glucose. She had a lot of life stressors, but what I'm showing you here are two consecutive days. And in the first day at seven weeks, six days, she's using manual mode for part of the time. And you can see she has a severe low and then she has a severe high. And when she's in auto mode consistently the next day, she has much less variability. Not a perfect tracing for pregnancy, but doing much better with automation. The other thing that hybrid close-up therapy can be beneficial for is quality of life. So in clip four, they actually did semi-structured interviews and women reported improvements in peace of mind, superior glycemic control. Related to CGM benefits, flexible lifestyle. There were a number of perceived benefits from a quality of life perspective by using hybrid closed loop in pregnancy. And then I also wanna show you that there are ways with other off-label assisted techniques with more current systems. So I have a patient, this is a real life case from my clinic. I haven't published it yet. It's unpublished data, but she came in to see me at four weeks, three days for her first pregnancy visit. Her A1C was 6.3% at the start of the pregnancy. She had tandem with control IQ technology. She was using control IQ only part of the time, about 60% of the time. Her average sensor glucose was 142 and her pregnancy specific time and range was 59% with 40% highs. What I did is I used some of these assisted techniques. I put her glucose target range in SLEEM mode, which is the lowest target range. We worked on basal bolus distribution. We worked on pre-mil bolusing. I changed her pump settings around and just two weeks later at six weeks of pregnancy, her average sensor glucose is 122 and her time and range is 78%. So again, in my opinion, in some women, we have the advantage that we can get that time and range above 70 early on with these technologies. And I'm very, very happy to report that currently there are five ongoing studies about hybrid closed-loop therapy in pregnancy. I just finished up my randomized control trial of the Medtronic 670G system. Helen Murphy just completed her pivotal trial for the CAM APS system, which is approved in Europe with a pregnancy specific algorithm and it has a CE mark approval. There's also a study with a customized pregnancy specific algorithm, Eyal Dessau and his group, and then studies with the tandem control IQ and Medtronic 780G. So more information to come on those. So in summary for my section about advantages, optimizing glucose control is hard, especially in type one diabetes pregnancies. Women don't always accomplish it on their own. Maybe closed-loop could help. Published completed trials are showing promise. Case reports of use are showing promise. Assistive techniques with commercially available products being used off-label can also be promising and there are new clinical trials underway. And now with that, I'll pass it on to my esteemed colleague, Dr. Vallant. Thank you so much. Thanks everybody for being here. I'm gonna go through some of the disadvantages. So, Eaton and I laugh because, you know, when we were debating different sides, but honestly our approach is very similar with these types of situations as things need to be individualized for patients. But without ado, certainly just as Dr. Polsky was saying, diabetes is so challenging to manage. And I think this is a nice example of a publication that was looking at pregnancy outcomes that was assessed prospectively among patients with type one diabetes and the diabetes and preeclampsia trial in Ireland. And basically just obviously demonstrating that majority of patients even in the third trimester when things, you know, or, you know, from I should say late second trimester going into third trimester, we really can't achieve the ranges that we need until very late in the pregnancy. But certainly glycemia matters not just at the end of pregnancy, but all throughout. And so if we're already having issues achieving these goals of what we need to be able to accomplish for our goals to be able to have good pregnancy outcomes both for the offspring and for the gestational carrier, then it's really important to think about, well, what are ways that we can actually be able to do to achieve these? And so this is just looking at A1C as the referent group of less than 6%. And then obviously demonstrating that with increasing glycemia, we have greater adverse rates of adverse outcomes. And in addition to that, of course, when we, the conversation here is looking at technology interventions that can help achieve more optimal glycemia ranges, but it's more complex than that. There's so much that the metabolic system, particularly in pregnancy is so sensitive to. Of course, we have to think about the fuels that we put into our body, the physical activity, of course, diabetes related complications, particularly in this population. We know that sleep and circadian rhythm and chrononutrition is significantly impacts how our bodies respond. And then of course, the social determinants of health and our ability for our patients to be able to actually achieve and be compliant with the care that we actually recommend. So I'm gonna actually start with my summary of disadvantages more than ending with them, more for the fact that of course, the pregnancy specific target ranges that we have doesn't matter if you have type one diabetes or gestational diabetes. And from a hybrid closed loop system, as the commercially available ones were discussed, do have some setting boundaries that don't allow us necessarily to be able to achieve some pregnancy specific, particularly in the United States. And so those are probably our biggest challenges that we'll kind of go through, but certainly other contributors are certainly meal timing and carbohydrate distribution. We'll talk a lot about that. And then as discussed before, looking at our significant insulin resistance curves that happen, we also have periods of insulin sensitivity where obviously that is also challenging to the system for us to be able to manage. And then these hybrid closed loop systems are completely dependent on people being able to actually make sure that they put in information about when they're eating. And prebolus is certainly very important in pregnancy to be able to make sure that we achieve the ranges that we need. And then of course, it's a feedback loop to be able to communicate between the different systems, to allow for the algorithm changes, and there's certainly delays and responses to the rises and fall, particularly in pregnancy. And then of course, we have a lot of people coming in now that are on hybrid closed loop who've never managed pump therapy in manual mode or otherwise, and therefore their overall understanding of pump technology may not be to the point where they're able to actually troubleshoot through technical challenges. So pregnancy targets. I think this is a huge controversial area within the field of obstetrics, more for the fact that the HAPO trial is a huge challenge for the world for us to understand that our risks for adverse pregnancy outcomes certainly increase at much lower glycemia ranges than we had previously appreciated. And so this is just obviously demonstrating the main outcomes of the HAPO trial, looking at large for gestational age, primary cesarean rates, neonatal hypoglycemia, and cord blood C-peptide. But again, just demonstrating the fact that we're anticipating blood sugars right now, assuming that that is the primary glucocentric focus of intervention, that where we're anticipating our ranges to be for optimal pregnancy outcomes may need to be further looked at. ACOG and American Diabetes Association have our pregnancy targets as such, as everybody's aware. But if we actually look at patterns of euglycemia, and this was the study that was looked at by Terry Hernandez et al, looking at pregnant persons without diabetes in pregnancy and kind of looking what their overall blood sugars look like. And really it's a very tight glycemic range at much lower levels, kind of similar to what the HAPO trial is suggesting. And so really our mean targets of having an average mean of 100 generally, are those really appropriate for us to be able to achieve patterns of euglycemia as those pregnancies don't have as similar rates of adverse pregnancy outcomes. So what about hybrid closed loop though? What can they actually achieve of what's actually on the market and what our patients have available to us? Because of course, if we can have interventions that can help achieve improved targets, that can be several years from now, or it could be a long ways away, or it could be very short period. But at least what do we have to our advantages right now? And certainly there's limitations as our glucose targets in these hybrid closed loop systems that are commercially available are very limited for what we can target down to. So again, focusing on the glycemic targets of what ACOG, ADA, and then on top of that, if we really want to be able to do it in a very dynamic period, such as labor or delivery, we have even more strict glycemic targets in those situations. And the type one diabetes data really drives that labor data for us to know that even just incremental increases in glucose by five milligrams per deciliter can increase the risk for neonatal hypoglycemia even within the context of labor. So certainly glycemia is really important. Our targets are strict in pregnancy and for what interventions are truly able to achieve that at this point is kind of part of the conversation here. So if I present the similar, the same slide that Dr. Polsky provided really are, I bolded here, the mean glucose ranges. We are far from target whether we're using sensor augmented or certainly hybrid closed loop. And so whether or not we're ready to use this in pregnancy prime time, I think is very individualized, but certainly something to consider, but also something to be in caution with if people are not comfortable using it or if patients aren't super comfortable using it. This was part of CLIP-3 and CLIP-4 trials. They wanted to look exploratory into the labor and delivery realm of their patients. And so labor was considered 24 hours before delivery. The median gestational age of delivery was 37 weeks, but you can see here that the mean glucose was between 113 and 126 in this group. They had high rates of neonatal hypoglycemia and similar percentages that still went to the NICU. Similarly, if we look at the same population of people that Dr. Polsky pointed out, some good cases or individuals that did less well on sensor augmented therapy, we can also find individuals that did better on sensor augmented therapy also. So it's not to say that hybrid closed loop, it should necessarily not be used, but more so that we should be individualizing it by the patient. I think one of the major areas that cannot necessarily be ignored is the fact that pregnancy is pretty complicated. There's lots of circulating concentrations of glucoregulatory hormones that are significantly altered in pregnancy that certainly challenge the system. And it's not to say that these cannot be considered within the algorithms, but these are not consistent throughout the day, of course. The other aspect of it too is for the fact that the nutritional demands of what the placenta and the fetus needs are primarily glucose. So we can't necessarily negate the fact that nutrition plays a large role in our ability for hybrid closed loop therapy to work well, but really in any sort of diabetes management to work well. This figure here is showing that the fuel demands primarily looking at the brain, which prefers glucose, but also the developing fetus and the placenta is often the forgotten organ that also primarily prefers glucose as the primary fuel. So when we think about glucose from that perspective, looking at how do we even get to the RDA of 175 milligrams, 175 milligrams, Dr. Hernandez out in Colorado and has kind of taken this and we've all talked about it, but certainly because the placenta is such a large glucose consumer, we haven't really thought about the capacity of that as well. And she's taken the four vessel data and been able to extrapolate potentially how much placental glucose consumption is and really our overall RDA of how much carbohydrates we should be consuming maybe even actually higher than what we actually recommend right now. And so in thinking about this, we know that the whole point of glycemic control, of course, is dependent on the consumption of fuel also not just our basal rates. And so thinking about how this needs to be incorporated and how to be able to address this is really important as well. Of course, we have to respect the increasing insulin demand and then understanding also that the overall insulin absorption is delayed as we advance in gestation. So the pre-bolus is not just only an important factor for glucose control perennially, but also the timing of pre-bolusing is also really important. So we can have a hybrid closed loop system, but if it's dependent on the user to be able to pre-bolus and really think about where they are in pregnancy and how their body is actually responding to the insulin absorption, we have to be mindful of how that's going to impact the ability for hybrid closed loop to actually work effectively. And then we also, for all of us taking care of patients with diabetes and pregnancy, we know how it's not just the carbs that we consume. There's other aspects of macronutrients that can influence how our carbs and our bodies respond to carbs. So we can't ignore the fact that our general consumption in pregnancy and outside of pregnancy for sure, but our general consumption of our added sugar, saturated fats are high. And so we really need to understand what our patients are doing, how our patients are eating to be able to actually optimize the systems that we have, including hybrid closed loop. This was an interesting study that was recently published. It was a retrospective cohort study looking at patients with type one diabetes and looking at patients who initially had sensor augmented therapy in a prior pregnancy and then a subsequent pregnancy using completely hybrid closed loop only throughout the whole pregnancy and trying to identify what had achieved higher success of better glycemic control, particularly in the hybrid closed loop system. And it was really higher meal frequency. So again, going to the fact of how we're eating our foods and how that's contributing to our ability to be able to actually achieve better glycemic control is something to be considered strongly. This is just, of course, focusing only on the yellow box, more to say of what we should be consuming in pregnancy. It doesn't matter if we have diabetes or not, we should all be consuming healthy eating patterns with overall higher qualities of our macronutrients. And the difference with diabetes is really just meal frequency and carbohydrate distribution. If we look at the same CLIP4 study and looking at what people had to say as far as the disadvantages or what could be improved necessarily with hybrid closed loop therapy are true. More for the fact that there could be technical glitches, whether it's CGM or non-CGM related, the alarms, the algorithm, mostly, of course, thinking about leading to suboptimal control, mostly for the limitations and the targets. And then, of course, the other challenges of having loss of control based to technology. All right, we are gonna go on to some clinical pearls for commercially available hybrid closed loop therapies in pregnancy. I'll have Dr. Polsky start. Sorry. So first, some general recommendations. We encourage preconception counseling when possible for all women with preexisting diabetes and pregnancy counseling at the first visit. Do you want me to just do this whole slide, Amy, or, okay. And then you should counsel women on the individual components of a closed loop system. Every time a woman comes into my clinic and we talk about it, we talk about the components like the CGM and the pump and specifically things like, for example, with the pump, you're getting only short-acting insulin. So if there's an interruption in insulin delivery for any reason, the cannula is kinked, the tubing is crushed, the insulin reservoir is low, you have no insulin delivery whatsoever because you don't have a long-acting basal shot in your system, increased risk for DKA. So we have to be careful about that. In my opinion, anyone who's thinking about using closed loop in a woman with pregnancy should first get experience with that system outside of pregnancy. Become comfortable with the system in a non-pregnant population. And then when you feel comfortable, if you feel like that's appropriate for an individual, then you can consider using it in the individual. And then use caution when you introduce an insulin pump or hybrid closed loop therapy during pregnancy. I do believe that you can definitely start a woman on insulin pump during pregnancy safely with the appropriate educational and support staff. You need that staff on hand to help onboard them in a safe way. And then it's really important, as Dr. Vallant said, to adhere to pregnancy dietary and clinical care recommendations. I mean, my opinion, you know, if you're consuming 355 grams of carb per day, there's no system that's going to keep up with that super well. So we want to adhere to regular recommendations like 175 grams of digestible carb divided over the day and come to your clinical care visits on a regular basis to optimize therapy. Great. Thank you so much. And so, you know, when, since the commercially ones available, patients are coming in with it, we have to know how to be able to help our patients who want to continue using this and optimize as much as we can. And so certainly as in, hopefully all of our standards of care that we individualize therapy either a week per week, or sometimes multiple times per week, depending on where they are in pregnancy and how their sensitivity is. Certainly as some of the examples that Dr. Polsky showed before, the using the lowest target that's available to the system is optimal. And then particularly, you know, understanding kind of where the advantages of that particular patient is, you can kind of use it in that way to be advantageous. Some of the fake carbohydrates, unfortunately, is one of the realities in order to be able to achieve. And that takes a lot of time and that takes a lot of dedication. And so, you know, that might not be capable for all patients that come into practice, because it does take a lot of diligence in order to be able to keep those sugars under really good control. Of course, avoiding prolonged basal insulin suspensions. I'm sure you've had plenty of patients who've come to you to be like, I didn't even realize my, you know, my basal had suspended until hyperglycemic ranges were noticed. So trying to be diligent about that. And then of course, the increased prebolus time over gestation as indicated. Everybody's so variable with that, but in particular, it can be up to like 45 minutes to an hour of prebolus in the third trimester. And then of course, the superbolus around mealtimes as needed, as indicated per patient. Dr. Patti. So some takeaway points about closed-loop pregnancy. Optimizing glycemic control on type 1 diabetes pregnancies is very challenging. Hybrid closed-loop therapy is still off-label, except for in Europe where CAM-APS has been approved, and it's still investigational in the U.S. There are published completed clinical trials that are small. They're limited in scope. They have short comparative durations, but they're showing promise. And there are case reports of commercially available systems, again, showing promise. And there are new clinical trials underway. So stay tuned. And then, you know, I wanna thank all the participants in my clinical trials and all the staff that I work with at the Barbara Davis Center and other facilities on my pregnancy studies. I learn from them every day and couldn't do the kind of work we do to try and improve the lives of our patients without their help. And at this point, Dr. Valen and I would be happy to take questions. Thank you so much, Dr. Polsky and Dr. Valen for a fantastic talk. I learned a lot, and I see that now a lot of questions are coming in. And also finishing up on time. So that's great. I think we have a really good time to take some questions. And I'm gonna start with this. We always get this question that are you planning to share your slides? Yeah, we will. Okay, so once Dr. Polsky and Dr. Valen provides a slide deck to us, probably we'll share that with you either through an interest group forum or some other way. Okay, other question. And I'm gonna start that with both of you actually, and probably Dr. Polsky, you can answer first followed by Dr. Valen. Is there any thoughts as why two increased glucose is related to significant increase in a preeclampsia? Yeah, that's a really good question. So preeclampsia is really a disorder of dysfunctional placentation. And the truth of the matter is that while glucose levels do affect the risk for preeclampsia, they're by no means the only thing that affect risk. So there are studies that are showing that women with a longer duration of diabetes, women with a history of any microangiopathy are a much greater risk. So if they already started off with retinopathy, with nephropathy, for example, their risk is greater. As to why glucose specifically might increase the risk further, I mean, it could be advanced glycation end products or other things, or it could be kind of a marker of poor overall health in a sense, again, leading to those microangiopathies to begin with. But I'd be really interested to hear what Dr. Valen thinks. Yeah, thanks. No, I echo a lot of what Dr. Polsky said. Certainly people with endothelial dysfunction are at higher risk for having abnormal placentation as that requires efficient angiogenesis to happen. Certainly glycemia is thought to also affect placentation more for the fact of some of the inflammation and some of the reactive oxygen species that are present more so when there's higher hyperglycemia surrounding obviously the intrauterine environment. Other things to consider, of course, too, is that bringing back to the fact that overall in early pregnancy, it's a relatively hypoxic environment, right? So it highly depends on a low resistance system. And when you're highly dependent on glucose, there's down regulation of some really important signaling markers, as well as, again, some of the other factors that you have from a fuel utilization to be able to actually develop that robust trophoactoderm into kind of the pathway into placentation. So hopefully that kind of helps. Thank you. I'm seeing that there are a lot of questions now, so I would appreciate if you can keep it a little bit quicker and shorter in your answers so that we can take more questions. So this is more about like a research question that despite that technologies are advancing, still we are seeing some poor outcomes, both maternal and fetal, right? Why are we just looking at the glucose? There can be other things that may be causing this bad outcomes, right? Like for example, the placenta itself or structure and genetics and so on. Anyway, I'm not up to date in that literature, but I'm going to start with you, Dr. Vailant. Any studies about this differences in a placental structure between type one, type two, gestational? Yeah, great. I'm going to keep this really brief. I agree. I think we're very glucocentric with diabetes. Certainly the phenotype of a type one is kind of mixing a little bit now that with increasing obesity, even in the type one population, but there's really good evidence to suggest even like particularly in type twos and GDMs, which is on the same spectrum of disease that lipid levels and differences in lipid biology can influence adverse outcomes as well. So I agree. I think that glucose is a achievable thing. We need to study other aspects, of course, that influence placental function, which is ultimately leading to a lot of these outcomes, but we need to be able to achieve even glucose. So, but you're right. I 100% agree. Okay, great. Thank you. I would just add to that that I totally agree, but I would just add to that that a caveat is that a lot of the research has a later age of enrollment and gestation. So like concept that mean age of enrollment was around 12 weeks after organogenesis. In the closed loop studies that I showed you, they had a two to four week run-in phase, four weeks maybe of SAPT and two week of washout. So really they're only using that closed loop for third trimester and part of second. So my personal opinion is that we should also be doing studies that do look to see if we control the glucose earlier, early in organogenesis, do we then affect these outcomes? And I think the answer would be yes. Okay, great. So now many clinical question. So I'm gonna start you with Dr. Polsky and I would be interested to also hear from Dr. Wayland on what you guys do on your side. So it is going to be a slightly rapid fire. So timing range 63 to 140, we know that current closed loop systems are suspending way before this 62, 63, which is considered as a normal at the time. And it's also quite possible that some people might have a 120 mean glucose and still a good timing range. And some people will have 80 mean glucose and still same timing range, right? So anyway, and none of the metrics are perfect. So the question here is that how do you really assess all these different metrics when you are managing patient with type one or type two or gestational diabetes during pregnancy? Yeah, I mean, I think you have to take the whole picture. I look at everything. I look at the mean sensor glucose, time below range, time within, time above. And I use that as a tool to explain why I'm making pump adjustments. So for example, a woman may have a good timing range, but she has very high hypoglycemia. And almost all the hyperglycemia is in an overreaction to the hypo. So yeah, your timing range is 71%. That's that goal. But you also have 8% time below range. If we focus on getting rid of these lows, we'll get rid of the highs. So I look at it as the whole picture and what is the glucose variability and moving forward that helps women, I feel, yeah. Okay, quickly, do you use GMI in pregnancy? Because when GMI, which is an estimated A1C formula from a linear regression was developed from a non-pregnant cohort. Yeah, good question. So Dr. Shah and I actually published a small report of 27 women that we followed. And GMI was a better correlator of the glucose picture in early pregnancy. The correlation between like time and range in A1C or GMI and time in A1C in second and third trimester was quite high. But particularly at the beginning when that A1C is being lowered at a very rapid pace, I find that GMI is actually more reliable indicator of our overall glucose control at that stage. And I use GMI all the time. Also telehealth, sometimes I don't have an A1C and so I have to rely on that GMI. With I'm sure that understanding its limitations, right? Okay, so this question is to, I would say both of you is that there are a lot of practical questions around, let's say if you use the 670G or tandem control IQ or Omnipot 5 during pregnancy, what kind of factors that are modifiable? What do you do to optimize the control, number one? And I'm combining a lot of questions into this big question. Number two, each trimester is different, right? So what is the practical consideration around those adjustments? And number three, what do you do? And this is probably, I'm really curious to know from Dr. Valente is that after labor, it's a whole different game, right? Like, so, and do you really put somebody on some kind of injections or do you really start directly to these automated insulin therapy? So anyway, it's a lot of questions around that part. So if you can spend a minute or two answering those practical question, I think audience will appreciate that. Really great questions. I have women on everything in my clinic, literally like pretty much every Medtronic pump, tandem, basal IQ, control IQ, DIY, Omnipot 5, everything. You have to know what is modifiable in each individual system. There's no way around it. So for example, if you change the basal rates in a Medtronic system, that doesn't affect the closed loop algorithm. So it's good to do that. So if they fall back into manual mode, they have some safe settings, but it doesn't affect their insulin delivery and automation. On the alternate side of things, in tandem, it does. So unfortunately, it's just one of those things that you have to know which system does what. The carb ratios are important for all of them. The correction factors are only important for some of them, the active insulin time for some of them. So it's just knowing each system individually. There was a question about what do you do in the different stages of pregnancy? So typically in the first trimester, and especially if a woman comes into the pregnancy with an A1C at goal, we're having to back off on her insulin doses, both basal and bolus. And then as insulin resistance ensues, we have to go up on both, but much more dramatically on the carbohydrate to insulin ratios and on the correction factors. And in fact, there is a discrepant and like over proportion of the bolus. And what I have found is that when someone's not in automation, that basal bolus distribution is roughly 30% basal, 70% basal in mid gestation onward. But when they're in closed loop, typically it works better with this 20, 80%, a split for their total insulin. Okay, so since you talked about a little bit about control IQ, sorry, I'm gonna ask you this question right now to you before how we'll take Dr. Whalen's answer on the labor and after delivery, is that do you prefer number one, sleep mode versus non-sleep mode on a control IQ? And number two is that, is there any particular individual cases that you prefer actually basal IQ rather than control IQ? That's a great question. I 100% individualize my therapy for each person. So if I know a woman and oftentimes I do, because we see women before, during and after their pregnancy in the same clinic, if I know a woman and I know that she's gonna do super well with just protection from lows, I'm gonna put her in basal IQ and maybe her average sense of glucose is 105 and maybe her time in range is 85%. So putting her in automation is the wrong choice for that person. If I know a woman and I know she struggles as it is, then control IQ may be the better option for her. Typically for almost all the women that we have in control IQ off-label, we put it in sleep mode, which has a target of 112 and a half to 120, 24 hours a day. So you program the sleep schedule. So it's from midnight to 1159 PM every day. But I tell them you are 100% responsible for every bolus. There is no automatic correction bolus through the pump, but that's good because if you have control IQ without sleep mode and just the regular mode, it will give you 60% of the bolus based on your correction factor. If it's predicted, you're gonna be within 180 within 30 minutes. That's too slow and it's also too little insulin. I want you to deliver a full bolus when you're high, not a partial bolus. Now, occasionally someone's like, never bolusing. So then we might not put it in sleep mode so they can get that automatic correction bolus in some way. Thank you. Dr. Vellant about paper and after that. Yeah, I agree with all the things that Dr. Polsky said. So those are like really great, as she calls it, assisted techniques, which are great. And I agree with the sleep mode. Like patients feel that they can at least have some sort of a sense of control over that. So that's also really good. From a labor and delivery perspective, honestly, like from a, during labor and delivery, because the glycemic targets are actually even stricter during labor and delivery, it's very, most patients when we do shared decision-making don't wanna stay in automation, to be honest. And there's also limitations based on like where they're delivering and stuff, of course, and the comfort and level and hospital policy. So you also have to consider those factors. But in postpartum though, I actually love hybrid closed-loop for patients that have it more for the fact that they're, as you saw in that insulin sensitivity curve, that they become very sensitive, particularly within the first two weeks of pregnancy as their mature milk comes in, if they're planning on breastfeeding or chestfeeding. And so that's a really, and we don't worry as much. There's no second passenger we have to worry about during that period. So really the goal is to keep that person safe. And so automation really provides excellent role for that. Yeah, we are actually one minute above our timeline, but I'm gonna take two quick questions. Dr. Polsky, what is this fake bolus? And how do you know that how much fake bolus is the right bolus for that individual? And what is super bolus? Yeah, okay, so fake carbs means you're putting carbs into the pump to deliver a bolus, but you're not actually consuming the carbohydrates. There is no one size fits all approach to this. What I do is I base the amount that I recommend on their stage of insulin sensitivity. So if it's, for example, someone has a high carbohydrate to insulin ratio, like 30 to one, that's a very insulin sensitive patient. So we're gonna put in fewer carbs for that person. Someone who's very resistant, has a carb ratio of two to one at the start of pregnancy, needs a lot more. So again, it's an individualized approach. For people kind of in the middle, which is most of our patients, roughly five to 20 grams for a fake carb bolus, or 0.5 to two units for a manual bolus, start conservatively, adjust as needed. This is also, by the way, advice that I've discussed with other colleagues, like Emily Smilovich, Liz Beshore, and Carol Levy, and we all agree on this approach. So I think that that's a good kind of one size fits all. The other question was about super bolus. This is actually a concept developed by John Walsh. And then I would say further defined by Lois Donovan. And the idea is around the time of a meal, especially if a woman has regular eating times, I find this technique to work super well. We increase the basal insulin delivery, and we ramp up on that carbohydrate to insulin ratio to make it more aggressive, so that we reduce the postprandial peak in their hyperglycemia. The problem with that, is now we have a lot of insulin on board, let's say between 6 and 9 a.m. for breakfast. And so we do reduce that peak, so it doesn't get to 140 postprandial, but then they get late hypoglycemia. So then we steal from the basal, or we take away from the basal, like two to three hours afterwards. So for example, I might have a 6 to 9 a.m. basal rate of 1.0, and then a 9 a.m. to 11 a.m. basal rate of 0.65. And the idea there is we are bolusing aggressively around the time of the meal, but we're protecting the woman from late postprandial hypoglycemia, like two to three hours later. And that's the concept of the super bolus. All right, thank you so much. I think we have to limit this webinar here, but thank you everyone who joined this webinar. Thank you so much, Dr. Polsky and Dr. Welland for a fabulous presentation. A lot to learn, and I understand this area is evolving, as you showed that there are many studies going on, and I'm hoping that it will give us more answer in future. There are a lot of questions that we could not answer live at this moment, but Dr. Polsky and Dr. Welland shared their emails, so feel free to email them if there is a particular or individual clinical questions that you would like them to answer. Again, thank you everyone for sharing your hour with us. I hope this was really a good webinar for you. For more webinars, again, go to the Professional ADA website and register for similar webinars like this one to learn more about use of a technology in improving outcomes for our patients with diabetes. With that, thank you so much, and bye-bye for now. Thank you, bye.

hybrid closed-loop therapy

pregnancies

diabetes

glucose levels

closed-loop systems

individualized therapy

preconception counseling

glycemic control

pregnancy dietary care

clinical care recommendations